{"id":"W3159409148","doi":"10.1093/database/baab021","title":"Increasing metadata coverage of SRA BioSample entries using deep learning–based named entity recognition","year":2021,"lang":"en","type":"article","venue":"Database","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institutes of Health Research; National Institutes of Health; National Institute of General Medical Sciences; Canadian Institute for Advanced Research","keywords":"Metadata; Computer science; Named-entity recognition; Information retrieval; Artificial intelligence; Deep learning; Natural language processing; World Wide Web","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003590198,0.0001142758,0.0001739447,0.00004137738,0.00009522476,0.00003576672,0.0001178678,0.000108005,0.0001135934],"category_scores_gemma":[0.002468236,0.000113622,0.00007026361,0.0001440897,0.0001452039,0.00001291736,0.0001960089,0.0001122329,0.000004237736],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001063436,"about_ca_system_score_gemma":0.0001287404,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002492975,"about_ca_topic_score_gemma":0.00009793047,"domain_scores_codex":[0.9989018,0.0002444635,0.0002075212,0.000311036,0.0001483722,0.0001867892],"domain_scores_gemma":[0.999254,0.00008700849,0.0001178453,0.0003518255,0.0001191081,0.00007021322],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001866111,0.0001686676,0.009135325,0.0001171153,0.00009304475,0.00004246615,0.00002449843,0.00003492086,0.9619601,0.00001247654,0.0002500227,0.02797479],"study_design_scores_gemma":[0.0009974871,0.0001184958,0.001181527,0.00009198205,0.000114298,0.000050709,0.0002219284,0.001384381,0.9294215,0.00004089712,0.06611771,0.0002590358],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9088464,0.001329449,0.08848692,0.00004802977,0.0001124458,0.000051572,0.001059722,0.00001853222,0.00004693015],"genre_scores_gemma":[0.9490924,0.0003074819,0.03483543,0.0001803405,0.00008935053,0.000003614717,0.0154378,0.00001352496,0.00004001995],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.06586768,"threshold_uncertainty_score":0.4633372,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0404997178453738,"score_gpt":0.2904007959088666,"score_spread":0.2499010780634928,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}